Given a PDF file which contains scanned "split-page" views of each pair of pages, this script helps you reformat the document into one which contains each page as a separate one.

pdf_page_unsplit.py

# MIT License, Kristiyan Stoimenov 2023
#
# Use `--help` for a nice overview of the options.

import PyPDF2
import argparse

argparser = argparse.ArgumentParser()
argparser.add_argument("-o", "--output_file", help="output file")
argparser.add_argument("-i", "--input_file", help="input file")
argparser.add_argument('-s', '--skip-first-page', action='store_true', help='skip first page of the input file', default=False)

args = argparser.parse_args()

pdf_file = open(args.input_file, 'rb')
pdf_reader = PyPDF2.PdfReader(pdf_file)

pdf_writer = PyPDF2.PdfWriter()
for index, page in enumerate(pdf_reader.pages):
	(page_width, page_height) = page.mediabox.upper_right

	page.mediabox.right = int(page_width / 2)
	page.mediabox.left = 0
	if not (args.skip_first_page and index == 0):
		pdf_writer.add_page(page)

	page.mediabox.left = int(page_width / 2)
	page.mediabox.right = int(page_width)
	pdf_writer.add_page(page)

with open(args.output_file, 'wb') as output_file:
	pdf_writer.write(output_file)
	pdf_file.close()